ABSTRACT The prevalence of non-alcoholic fatty liver disease (NAFLD) in children has almost tripled over the past 20 years. NAFLD currently affects 8-12% of the general pediatric population in the U.S. and more that 30% of obese children. It is associated with an increased risk of developing advance stages of liver disease as well as cardiovascular and metabolic diseases. Mounting evidence suggests that early life environmental exposures contribute to the etiology of NAFLD. PFAS are persistent compounds widely used in water repellant textiles, nonstick coatings, and food packaging products, and have long half-lives (up to a decade) in humans. Almost all U.S. children and adolescents have detectable PFAS blood levels. Even low dose exposure to PFAS induces hepatotoxic effects in animal models. Despite abundant evidence from experimental studies, epidemiologic study is limited to a few cross-sectional studies in adults. We therefore propose a novel study design for investigating PFAS hepatotoxic effects in humans. We will leverage clinical and liver histopathological data from the Teen- Longitudinal Assessment of Bariatric Surgery (Teen-LABS) study, which is the largest national multi-center longitudinal, prospective study on teenagers undergoing bariatric surgery, and offers a unique archive of liver tissue and blood samples. We hypothesize that higher PFAS concentrations will be associated with NAFLD and non-alcoholic steatohepatitis (NASH, more severe NAFLD) at the time of surgery; furthermore, the large metabolic changes occurring after the bariatric surgery ?natural experiment? will magnify effects of PFAS exposures, resulting in attenuated improvement in liver injury after surgery. To test this hypothesis, we will use archived samples collected at the time of surgery to measure PFAS concentrations in plasma and liver and assess associations with liver histopathology at the time of surgery and with improvement in liver injury during follow up (Aims1&2). We will then identify pathways altered by PFAS exposure based on high resolution metabolomics profiles in liver tissue and plasma samples, using a hierarchical modeling approach (Aim 3). Finally, we will integrate results from the PFAS-omics analyses, using a novel latent variable modeling framework, to identify subgroups of adolescents who have less improvement in liver injury after bariatric surgery, based on their PFAS exposure and metabolomics profiles (Aim 4). The proposed research will be the first human study to examine the effects of PFAS exposure on NAFLD using the gold standard of liver biopsies for disease diagnosis and liver-specific and plasma metabolomic measures for examining biological mechanisms linking exposure to disease. A strong interdisciplinary team of investigators brings expertise in environmental epidemiology, pediatric hepatology, bariatric surgery, metabolomics, and biostatistics. The study, utilizing existing data and biosamples from a well-phenotyped clinical adolescent bariatric surgery cohort, is an innovative, cost-effective approach to advance our understanding of environmental contributions to pediatric liver disease that may identify new targets for prevention and intervention starting early in life.